ABSTRACT Hispanics/Latinos (Latinos) are the 2nd largest and fastest growing ethnic group in the United States. Although typically grouped as a single ethnic minority, Latinos are a heterogeneous population with diverse national origins, unique genetic admixture patterns (African, European, and Indigenous American ancestry), and a wide spectrum of socio-demographic characteristics. Colorectal cancer (CRC) is the second most common and fatal cancer among all Latinos combined; however, mortality rates differ substantially within Latino subpopulations defined, thus far, by place of birth. Rates are substantially higher in Puerto Ricans and Cubans, the subgroups with the highest African ancestry, than in Mexicans and non-Hispanic Whites (NHW). After accounting for socioeconomic status and access to care inequalities, differences in mortality persist, highlighting the need to accurately characterize and critically assess biological contributors to intra- and inter-ethnic group disparities. To overcome Latino underrepresentation in publicly available datasets, we propose to join resources from well- annotated epidemiologic studies including the Hispanic Colorectal Cancer Study (HCCS, R01CA155101), the Puerto Rico Biobank (PRBB, U54CA163068), and the Moffitt Cancer Center Total Cancer Care (TCC) cohort into a new consortium, the Latino Colorectal Cancer Consortium (LC3). This strategy allows us to maximize sample size and adequately represent the most common Latino subgroups in the US (i.e. Mexican, with high Indigenous American ancestry; Puerto Rican; and Cuban). We will characterize the somatic mutational profiles of Latino CRCs by whole exome sequencing using previously-collected tissues from the HCCS, PRBB, and TCC (n=600). For comparisons across racial/ethnic groups, we will leverage existing tumor sequencing data from NHW and African American (AA) CRCs in the Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO; 7,000 NHW), The Cancer Genome Atlas (283 NHW and 61 AA), and TCC (589 NHW and 40 AA). Here, we will examine driver mutation frequencies associated with estimated local and global genetic ancestry in Latino CRC cases (Aim 1), conduct a trans-ethnic analysis comparing frequencies of ancestry-associated and known clinically-actionable driver mutations in Latinos to NHW and AA (Aim 2), and assess the implications of genetic ancestry for survival as well as differential efficacy of conventional and potential targeted therapies for metastatic CRC using patient-derived xenograft (PDX)-based ex vivo live tissue sensitivity assays (LTSA) (Aim 3). Ancestry-specific PDX models will be established using fresh tissue from 40 prospectively recruited Latino, AA, and NHW TCC participants. Our innovative LTSA approach permits timely screening of multiple therapeutic agents while maintaining the tumor microenvironment and reliably predicting clinical responses. Results from this study will help us better understand the biological underpinnings of outcome disparities in Latino subgroups to inform translational efforts towards precision clinical applications.